C-SWAT: A Modified Revision of SWAT Using Consolidated Input Files

Haw Yen1,2, Mehdi Ahmadi3, Michael J. White1

Xiuying Wang2,Jeffrey G. Arnold1

Javier M. Osorio Leyton2

1 Grassland, Soil & Water Research Laboratory, USDA-ARS2 Blackland Research & Extension Center, Texas A&M University3 Department of Civil and Environmental Engineering, Colorado State University

June 25, 2015International SWAT Conference 2015Pula, Sardinia, Italy

Overview

Soil and Water Assessment Tool (SWAT)

General Calibration Process

Structure (format) of SWAT Input Files

Case Study & Results

Discussion & Conclusion

Outline

Development of complex watershed models

◉ Evaluate impact from climate changing, various human activities on issues

Advanced technology in computer science

◉ Complex watershed simulation models

◉ Distributed in space & process-based

◉ Long term simulations with large amount of input data

◉ High resolution input datasets

Iterative processes

SWAT discretization (Basin-Subbasin-HRU)

◉ SWAT with 10K HRUs = 2hrs ; 1K iterations = 8d

Overview

Modify SWAT input structure to consolidate HRU and subbasins into single file for each category to enhance the computational speed of SWAT model.

Objectives

Structure (format) of SWAT Input Files

SWAT Input Files

HRU Level File Structure Group

*.chm II

*.gw I

*.hru I

*.mgt I & III

*.ops IV

*.sep I

*.sol I & II

Subbasin Level File Structure Group

*.pnd I

*.rte I & II

*.sub I & II

*.swq I

*.wgn I & II

*.wus II

1

Structure I

Each row contains only one parameter value

Structure II

Each row contains more than one parameter value

Structure III

Format in MGT files

Structure IV

Format in OPS files

Little Washita River Basin

◉ Oklahoma, USA

◉ 611 km2

Case Study Area

Case Study Area

◉ Normal SWAT project

◉ 75 Subbasins

◉ 394 hru

◉ Total 3148

◉ C-SWAT Project

◉ 7 hru

◉ 6 Subbasin

◉ Total 43 files

◉ 98% fewer files

General Calibration Process

Comparison of runtime spent on each conducted calibration

Results (1/2)

Application of parallel computation

Results (2/2)

C-SWAT is no doubt a time/effort saver

◉ Further improvement can be made by applying C-SWAT on large-scale watershed projects

◉ Source code available (Haw Yen)

◉ C-SWAT can be applied on other revisions

The upcoming New Generation SWAT (modular code) will adopt the concept of C-SWAT

◉ Input files will be consolidated

◉ More associated modifications

Discussion and Conclusion

C-SWAT Theory & Development◉ Yen, H., M. Ahmadi, M. J. White, X. Wang, J. G. Arnold (2014) “C-SWAT:

The Soil and Water Assessment Tool with Consolidated Input Files in Alleviating Computational Burden of Recursive Simulations.” Computers & Geosciences, 72, pp. 221-232.

More Implementations◉ Yen, H., X. Wang, D. G. Fontane, M. Arabi, R. D. Harmel (2014) “A

Framework for Propagation of Uncertainty Contributed by Input Data, Parameterization, Model Structure, and Calibration/Validation Data in Watershed Modeling.” Environmental Modelling and Software, 54, pp. 211-221.

◉ Yen, H., R. T. Bailey, M. Arabi, M. Ahmadi, M. J. White, J. G. Arnold (2014) “The Role of Interior Watershed Processes in Improving Parameter Estimation and Performance of Watershed Models.” Journal of Environmental Quality, 43(5), pp. 1601-1613.

Reference

This study was funded by:

◉ US Department of Agriculture– National Resources Conservation Service

(USDA-NRCS) Conservation Effects Assessment Project (CEAP) – Wildlife

and Cropland components.

◉ International S&T Cooperation Program from the Ministry of Science and

Technology of China (2012DFA91530)

◉ National Natural Science Foundation of China (41161140353, 91325302)

◉ First Youth Excellent Talents Program of the Organization Department of the

Central Committee of the CPC

◉ Fundamental Research Funds for the Central Universities (TD-JC-2013-2)

Please do not forget that USDA is an equal opportunity employer and

provider!

Acknowledgement

Thanks for your attention!

Haw Yen, Ph.D.

Assistant Research Scientist

hyen@brc.tamus.edu